New data from the National Institutes of Health–funded Osteoarthritis Initiative suggest that, in some women at least, taking bisphosphonates may help to reduce the chances that there will be radiographic progression of knee osteoarthritis (OA).
In a propensity-matched cohort analysis, women who had a Kellgren and Lawrence (KL) grade of less than 2 and who used bisphosphonates were half as likely as those who did not use bisphosphonates to have radiographic OA progression at 2 years (hazard ratio, 0.53; 95% confidence interval, 0.35-0.79). Radiographic OA progression has been defined as a one-step increase in the KL grade.
While the association appeared even stronger in women with a KL grade less than 2 and who were not overweight (HR, 0.49; 95% CI, 0.26-0.92), bisphosphonate use was not associated with radiographic OA progression in women with a higher (≥2) KL grade (HR, 1.06; 95% CI, 0.83-1.35).
“In all analyses, the effect of bisphosphonates was larger in radiographic-disease-naive individuals, suggesting protection using bisphosphonates may be more profound in those who do not already have evidence of knee damage or who have mild disease, and once damage occurs, bisphosphonate use may not have much effect,” Journal of Bone and Mineral Research., of the University of Toronto and her coauthors reported in the
“Our study was the first to our knowledge to examine bisphosphonate exposure effects in different disease severity subgroups and obesity classifications using a rigorous, propensity-matched time-to-event analysis that uniquely addresses confounding by indication,” Dr. Hayes and her team wrote.
Furthermore, they noted that extensive sensitivity analyses, which included redoing the primary analyses to look at statin use, showed that their main conclusions were unchanged and that this helped account for any potential residual confounding, healthy-user bias, or exposure misclassification.
The Osteoarthritis Initiative is a 10-year longitudinal cohort study conducted at four clinical sites in the United States and recruited men and women aged 45-75 years over a 2-year period starting in 2004. Dr. Hayes and her coauthors restricted their analyses to women 50 years and older. Their study population consisted of 344 bisphosphonate users and 344 bisphosphonate nonusers.
The main bisphosphonate being taken was alendronate (69%), and the average duration of bisphosphonate use was 3.3 years, but no significant effect of duration of use on radiographic progression was found.
The women were followed until the first radiographic OA progression, or the first missed visit or end of the 2-year follow-up period.
Overall, 95 (13.8%) of the 688 women included in the analysis experienced radiographic OA progression. Of those, 27 (3.9%) had a KL grade of less than 2 and 68 (9.8%) had a KL grade of 2 or greater. Ten women with KL less than 2 and 27 women with KL or 2 or greater were taking bisphosphonates at their baseline visit.
“Kaplan-Meier analysis indicated that non-users and users with a baseline KL grade of 0 or 1 had 2-year risks of progression of 10.5% and 5.9%, respectively, whereas non-users and users with a baseline KL grade of 2 or 3 had 2-year of these women risks of progression of 23.0% and 23.5%, respectively,” reported the authors.
Before propensity score matching, Dr. Hayes and her colleagues observed that women taking bisphosphonates were older, had lower body weight and a higher prevalence of any fracture or hip and vertebral fractures, and were also more likely be White, compared with non-users. “In addition, bisphosphonate-users appeared to be healthier than non-users, as suggested by a lower smoking prevalence, lower average baseline KL grade, lower diabetes prevalence, and higher multivitamin use (a healthy-user proxy),” they acknowledged.
Results in perspective
“The key thing that I’m concerned about when I see something like bisphosphonates and osteoarthritis is just how well confounding has been addressed,” commented, professor of medicine and epidemiology at Boston University and chief of rheumatology at Boston Medical Center, in an interview.
“So are there factors other than the bisphosphonates themselves that might explain the findings? It looks like they’ve taken into account a lot of important things that one would consider for trying to get the two groups to look as similar as possible,” she added. Dr. Neogi queried, however, if body mass index had been suitably been adjusted for even after propensity score matching.
“The effect estimate is quite large, so I do think there is some confounding. So I would feel comfortable saying that there’s a signal here for bisphosphonates in reducing the risk of progression among those who do not have radiographic OA at baseline,” Dr. Neogi observed.
“The context of all this is that there have been large, well-designed, randomized control trials of oral bisphosphonates from years ago that did not find any benefit of bisphosphonates in [terms of] radiographic OA progression,” Dr. Neogi explained.
In the, now considered “quite a large landmark study,” the efficacy of risedronate in providing symptom relief and slowing disease progression was studied in almost 2,500 patients. “They saw some improvements in signs and symptoms, but risedronate did not significantly reduce radiographic progression. [However] there were some signals on biomarkers,” Dr. Neogi said.
One of the issues is that radiographs are too insensitive to pick up early bone changes in OA, a fact not missed by Dr. Hayes et al. More recent research has thus looked to using more sensitive imaging methods, such as CT and MRI, such as a recent studylooking at the use of intravenous zoledronic acid on bone marrow lesions and cartilage volume. The results did not show any benefit of bisphosphonate use over 2 years.
“So even though we thought the MRI might provide a better way to detect a signal, it hasn’t panned out,” Dr. Neogi said.
But that’s not to say that there isn’t still a signal. Dr. Neogi’s most recent research has been using MRI to look at bone marrow lesion volume in women who were newly starting bisphosphonate therapy versus those who were not, and this has been just been accepted for publication.
“We found no difference in bone marrow lesion volume between the two groups. But in the women who had bone marrow lesions at baseline, there was a statistically significant greater proportion of women on bisphosphonates having a decrease in bone marrow lesion volume than the non-initiators,” she said.
So is there evidence that putting more women on bisphosphonates could prevent OA? “I’m not sure that you would be able to say that this should be something that all postmenopausal women should be on,” Dr. Neogi said.
“There’s a theoretical risk that has not been formally studied that, if you diminish bone turnover and you get more and more mineralization occurring, the bone potentially may have altered mechanical properties, become stiffer and, over the long term, that might not be good for OA.”
She added that, if there is already a clear clinical indication for bisphosphonate use, however, such as older women who have had a fracture and who should be on a bisphosphonate anyway, then “a bisphosphonate has the theoretical potential additional benefit for their osteoarthritis.”
The authors and Dr. Neogi had no conflicts of interest or relationships to disclose.
SOURCE: Hayes KN et al. J Bone Miner Res. 2020 July 14. .